The LTfLL project ('Language Technologies for Lifelong Learning') will create next-generation support and advice services to enhance individual and collaborative building of competences and knowledge creation in educational and organizational settings. The project makes extensive use of language technologies and cognitive models in the services.
The research activities are enveloped by activities that ensure common ground in use cases and pedagogically sound scenarios that steer the design and development of the services and guide the validation; a technical infrastructure for the creation and integration of the services and a validation structure that ensures rigorous evaluation in realistic settings, with several languages supported.
The research in the project is organized in 3 themes, each leading to particular types of services and infrastructures:
In theme 1 services are developed to establish the current position of the learner in a domain. Services will offer semi-automatic analysis and comparison of learner portfolios to the domain knowledge and continuous modelling and measurement of conceptual development.
In theme 2 support and feedback services are developed based on analysis of the interactions of students - using Natural Language Processing (NLP) and Social Network Analysis (SNA) – and textual output of students - using Latent Semantic Analysis with contributions from NLP.
In theme 3 a knowledge sharing infrastructure is construed that allows comparison and sharing of private knowledge to give rise to new common knowledge and social learning. Ontologies for formal domain representation are combined with social tagging.
The services are expected to result in improved appreciation of learner requirements, leading to better recommendations on study plans and resources. Progress monitoring based on learning activities, rather than on formal assessments, will improve recommendations for further competence building and improved co-construction of knowledge in social and informal learning.
Please, refer to the LTfLL project website for up-to-date information or contact Bernhard Hoisl.